vacoal-hippocampal-memory

star 2

VaCoAl algebro-deterministic hippocampal memory architecture. Bridges silicon and hippocampus by connecting Vector-HaSH, TEM, and hyperdimensional computing. Use when: hippocampal memory modeling, grid-cell scaffolds, episodic memory replay, sharp-wave ripples, Vector-HaSH, Tolman-Eichenbaum Machine, hyperdimensional computing, Galois-field LFSRs, multi-hop memory replay, memory fidelity decay, STDP-like path selection, EC-CA3 pathways, dentate gyrus, compositional memory, hippocampal-entorhinal circuit.

hiyenwong By hiyenwong schedule Updated 6/3/2026

name: vacoal-hippocampal-memory description: "VaCoAl algebro-deterministic hippocampal memory architecture. Bridges silicon and hippocampus by connecting Vector-HaSH, TEM, and hyperdimensional computing. Use when: hippocampal memory modeling, grid-cell scaffolds, episodic memory replay, sharp-wave ripples, Vector-HaSH, Tolman-Eichenbaum Machine, hyperdimensional computing, Galois-field LFSRs, multi-hop memory replay, memory fidelity decay, STDP-like path selection, EC-CA3 pathways, dentate gyrus, compositional memory, hippocampal-entorhinal circuit."

VaCoAl: Algebro-Deterministic Hippocampal Memory Architecture

Paper Reference

  • Title: Bridging Silicon and the Hippocampus: Algebro-Deterministic Memory "VaCoAl" as a Substrate for Vector-HaSH and TEM
  • arXiv: 2605.15652
  • Authors: Hiroyuki Chuma, Kanji Otsuka, Yoichi Sato
  • Date: 2026-05-19
  • Category: cs.NE (Neural and Evolutionary Computing)

Core Innovation

VaCoAl provides a shared algebraic object connecting three parallel research literatures:

  1. Vector-HaSH: Hippocampal-entorhinal circuit factorizes content from grid-cell scaffold
  2. TEM (Tolman-Eichenbaum Machine): Compositional memory via ripple-mediated replay
  3. Human iEEG findings: Sharp-wave ripples gate episodic recall; multi-hop replay fidelity decays multiplicatively

Key Components

1. Galois-Field Diffusion Substrate

  • Deterministic Galois-field LFSRs provide quasi-orthogonal vectors
  • Replaces Vector-HaSH's random scaffold-to-hippocampus projection
  • Advantages: Matched second-moment statistics, stronger avalanche behavior, bit-exact reproducibility
  • Architecture: Built from Linear Feedback Shift Registers over GF(2)

2. Confidence Ratio (CR) Model for Multi-Hop Replay

  • CR1: Per-step confidence ratio for single-hop reactivation
  • CR2: Path-integral = product of CR1 values along n-hop chain
  • Models observed multiplicative decay in multi-hop replay fidelity
  • First algebraically tractable model under conditional independence assumption

3. STDP-like Path Selection

Emerges from architectural demands:

  • Similarity preservation
  • Compositional reversibility
  • Bounded-frontier search

4. Hippocampal Pathway Correspondences

VaCoAl operating regimes share architectural commitments with:

  • EC → CA3: Direct entorhinal-to-CA3 pathway
  • EC → DG → CA3: Dentate gyrus-mediated pathway
  • Conserved across >520 Myr of evolution

Testable Predictions

  1. Multiplicative fidelity decay follows CR1^n form in iEEG recordings
  2. Ripple-locked cortical reactivation patterns match VaCoAl replay sequences
  3. DG elaboration in primates reflects capacity-plasticity tradeoff predicted by VaCoAl

Activation Keywords

hippocampal memory, VaCoAl, Vector-HaSH, Tolman-Eichenbaum Machine, grid cells, episodic memory, sharp-wave ripples, hyperdimensional computing, Galois field, LFSR, multi-hop replay, STDP, entorhinal cortex, CA3, dentate gyrus, compositional memory, memory fidelity decay

Install via CLI
npx skills add https://github.com/hiyenwong/ai_collection --skill vacoal-hippocampal-memory
Repository Details
star Stars 2
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator